Prenatal Testing
Identifying Drug Resistant TB
Breast Cancer Testing
Risk Factor Profiling

Prenatal Testing

Prenatal genetic testing has been recorded and researched nearly two decades. At first its use was limited to identifying gender, trisomy 21 and RhD. Recently, focus has turned to using genetic markers to monitor development, noting any abnormalities that may have consequences when the baby is born and later in life#.
SMART Genomics API supports capturing the type of sample (prenatal, somatic, or germline ) used in sequencing test in both Observation resource extended with SequencingLab extension and Sequence resource.
For example, API call below searches for sequences from infants
GET /Sequence?sampleType=prenatal&coordinates={...}
This feature will be useful in by allowing incorporation of the panel of prenatal genetic test results into apps along with clinical information from both the mother and unborn child, family history, and doctor’s notes.

Identifying Drug Resistant TB

As the number and variety of mycobacterial infections increases, it will be important clinically as well as for drug development purposes, to be able to identify species at a rapid speed. Being able to identify disease-causing organisms versus non-pathogenic species is important for the outcome of the patient as well as for the study of overall disease pathogenesis. Isolation techniques as well as treatment regimens may also vary depending on the type of infection present.
There have been multiple stabs at developing molecular tests for mycobacterial infections, but a few key roadblocks have yet to be removed. For example, there are highly polymorphic regions of an rRNA gene, which is used in several of the currently available assays#. These assays were designed for overall TB diagnosis and do not interpret all of the information present in this RNA region#.
In order to fully categorized mycobacterial variants, it is necessary to develop high-density DNA probe arrays. A recent study reports a first application of this idea where the authors designed an array containing all 16S rRNA polymorphisms over a 200bp region of a mycobacterial database. This allows for accurate and specific diagnosis of the infection.
SMART Genomcis API allows the encapsulation of microbial mutation by capturing the species from which source of sequencing test is collected. API call below searches for sequences from Tuberculosis from patient 123
GET /Sequence?species=113861009&patient=patient/123
This will be particularly handy in this situation if it allowed the combination of this new diagnostic information with phylogenetic information, clinical observations and tests, and could synthesize the various diagnostic outcomes for a clinician or researcher’s use.

Breast Cancer Testing

Genetic testing as recently been adopted as one of the tools in breast cancer screening, in addition to the more traditional anatomical and historical methods. A major hurdle, and one that has slowed the progress and development of the inclusion of genetics into cancer screening, is the problem of heterogeneous data, and how to synthesize that array of newly available information. In addition, the problem of effectively communicating the new data to the treating physician has also hindered the utilization of genetic data in such scenario. This situation calls for a well-designed API for genomic reads to allow clean workflow from various data types to the best possible end user experience.
In an collaboration with one of the major clinical genetic sequencing vendors in United States, SMART Genomics builds an adapter that reads from existing data of that vendor. After SMART-enabling the database, clinical application developers are able to easily utilize genetic data and write applications without being concerned with the underlying data structure of the vendor. One of the applications scans sequencing data of patients and looks for variants of BRCA2 that have been associated with breast cancer. For instance, the application makes below call to request sequence mapped to where BRCA2 is located.
GET Sequence?coordinates=chr17:41196312-41277500
After identifying variants linked to breast cancer, together with family history of the patient, clinicians can order further screening to confirm diagnosis.

Risk Factor Profiling

Imagine you have access to a knowledge database of the relationship between human DNA variation and phenotypes, and you would like to access a patient’s genotypes and match his genotype against those in the knowledge base. Let’s say, in the database you found that having a substitution of nucleotide C at location 3308 in a patient’s mitochondria has a pathogenic effect on him getting colorectal cancer, and you want to check if the patient has this variant. In this case you would want to the following query against that patient’s Sequence resource:
GET /sequence?patient=patient/123&type=dna&coordinates=mt:3307-3308
If with the query result you find that the patient does have the variant, you might want to warn him about that.
What else?
In addition to finding variant of interest, you can also profile the existence of such variant together with its associated trait (in the case of this example, it’s colorectal cancer) by creating a GeneticObservation resource and insert it into the patient’s health record with following operation.
POST /observation
By doing so, not only you are finding out a potential risk of the patient, you are also documenting the risk and sharing it with the other health advisors who has access to the patient’s record and who would benefit from you discovery.